MOSAIC: Learning Unified Multi-Sensory Object Property Representations for Robot Learning via Interactive Perception

TL;DR

MOSAIC is a framework that enables robots to learn unified representations of object properties across visual, auditory, and haptic modalities, improving multi-sensory perception and task performance.

Contribution

It introduces a novel multi-sensory learning framework that distills knowledge from foundation models and aligns representations across multiple sensory modalities for robotics.

Findings

MOSAIC achieves competitive object categorization accuracy.

The framework excels in zero-shot object-fetching tasks.

Unified representations improve multi-sensory perception in robots.

Abstract

A holistic understanding of object properties across diverse sensory modalities (e.g., visual, audio, and haptic) is essential for tasks ranging from object categorization to complex manipulation. Drawing inspiration from cognitive science studies that emphasize the significance of multi-sensory integration in human perception, we introduce MOSAIC (Multimodal Object property learning with Self-Attention and Interactive Comprehension), a novel framework designed to facilitate the learning of unified multi-sensory object property representations. While it is undeniable that visual information plays a prominent role, we acknowledge that many fundamental object properties extend beyond the visual domain to encompass attributes like texture, mass distribution, or sounds, which significantly influence how we interact with objects. In MOSAIC, we leverage this profound insight by distilling knowledge from multimodal foundation models and aligning these representations not only across vision but also haptic and auditory sensory modalities. Through extensive experiments on a dataset where a humanoid robot interacts with 100 objects across 10 exploratory behaviors, we demonstrate the versatility of MOSAIC in two task families: object categorization and object-fetching tasks. Our results underscore the efficacy of MOSAIC's unified representations, showing competitive performance in category recognition through a simple linear probe setup and excelling in the fetch object task under zero-shot transfer conditions. This work pioneers the application of sensory grounding in foundation models for robotics, promising a significant leap in multi-sensory perception capabilities for autonomous systems. We have released the code, datasets, and additional results: https://github.com/gtatiya/MOSAIC.